This application seeks five years of support to develop, evaluate and fine-tune a new research education program (WU-STAGEM: Washington University Statistical Training in Addiction Genetics Methodology), to train postdoctoral level scientists (9 over the 5-year funding period). Trainees (program participants) will include both US and foreign nationals, drawn from backgrounds such as biostatistics, mathematics, quantitative psychology and statistical genetics. We will provide trainees with both (i) the expertise to develop, test and apply new statistical and computational models to address genetics-based research problems in addiction, and (ii) sufficient training to allow participants to work collaboratively in multidisciplinary teams alongside basic and clinical researchers engaged in addiction research. Program postdoctoral scientists will train under the guidance of mentoring teams composed of (a) clinical researchers in addiction, (b) basic scientists and, most critically, (c) researchers with pertinent statistical or computational modeling expertise, with the support of individually tailored formal coursework. They will work with addiction geneticists at the host institution and elsewhere (e.g., through the NIDA Genetics Consortium) to identify and conduct research in areas where unique and urgently needed contributions to the field of addiction genetics can be made, producing products such as novel computer software, novel algorithms or approaches to statistical genetic analysis, to advance the field of addiction genetics. These priority areas will include (varying according to trainee backgrounds and interests and expert input): 1) incorporation of bioinformatics data in the statistical analysis and interpretation of GWAS data, 2) challenges in the combination of GWAS data across data-sets characterized by differing sampling schemes and somewhat different phenotypic assessments, and 3) modeling of genotype x environment interaction effects, and incorporation of consideration of GxE effects in efforts at gene-discovery. We will take advantage of existing data sets such as those at the NIDA Center for Genetic studies; zork.wustl.edu/nida/ and at dbGap, as well as more extensive data-sets associated with local investigators, in this training effort. Through infrastructure development and continuing evaluation efforts, building upon outstanding existing institutional strengths, we will seek to develop a self-sustaining program, drawing upon the wide-ranging expertise at Washington University that can supply highly trained experts in statistical and computational modeling to the field of addiction genetics.